提出SpLPG框架，降低分布式图神经网络链接预测的通信开销。
Demystifying Distributed Training of Graph Neural Networks for Link Prediction
- 通过图稀疏化减少各节点间通信量。
- 在保持准确率的同时，通信开销降低约80%。
- 适合大规模图数据的高效分布式训练场景。
图神经网络（GNN）是解决图相关问题的强大工具。分布式GNN框架和系统提升了GNN的可扩展性并加速了训练过程，但大多数优化针对节点分类任务。其在链接预测任务上的表现尚未充分探索。本文揭示了当每个工作节点仅在其分配的子图上训练GNN而无法访问全图时，性能下降的主要原因：不仅来自图划分带来的信息丢失，还源于训练中负样本采样方式的问题。虽然共享完整图信息可恢复精度，但通信成本过高。为此，本文提出SpLPG，利用图稀疏化有效缓解性能下降，显著降低通信开销。在多个公开真实世界数据集上的实验表明，SpLPG将通信开销减少约80%，同时基本保持链接预测准确性。
原文摘要 · Abstract (English)
Graph neural networks (GNNs) are powerful tools for solving graph-related problems. Distributed GNN frameworks and systems enhance the scalability of GNNs and accelerate model training, yet most are optimized for node classification. Their performance on link prediction remains underexplored. This paper demystifies distributed training of GNNs for link prediction by investigating the issue of performance degradation when each worker trains a GNN on its assigned partitioned subgraph without having access to the entire graph. We discover that the main sources of the issue come from not only the information loss caused by graph partitioning but also the ways of drawing negative samples during model training. While sharing the complete graph information with each worker resolves the issue and preserves link prediction accuracy, it incurs a high communication cost. We propose SpLPG, which effectively leverages graph sparsification to mitigate the issue of performance degradation at a reduced communication cost. Experiment results on several public real-world datasets demonstrate the effectiveness of SpLPG, which reduces the communication overhead by up to about 80% while mostly preserving link prediction accuracy.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。